MimosaNet：一种防止模型窃取的非鲁棒神经网络
MimosaNet: An Unrobust Neural Network Preventing Model Stealing
摘要：
深度神经网络对学习的网络参数的微小扰动具有鲁棒性，并且它们的微小修改不会显著改变整体网络响应。这为模型窃取提供了空间，恶意攻击者可以窃取已经训练好的网络，修改权重并声称新网络是他自己的知识产权。在某些情况下，这可能会阻止网络在嵌入式域中的自由分发和应用。在本文中，我们提出了一种方法，用于创建一个已经训练好的完全连接的深度神经网络的等效版本，该网络可以防止网络窃取：即它产生相同的响应和分类精度，但它对权重变化非常敏感。
英文摘要：
Deep Neural Networks are robust to minor perturbations of the learned network parameters and their minor modifications do not change the overall network response significantly. This allows space for model stealing, where a malevolent attacker can steal an already trained network, modify the weights and claim the new network his own intellectual property. In certain cases this can prevent the free distribution and application of networks in the embedded domain. In this paper, we propose a method for creating an equivalent version of an already trained fully connected deep neural network that can prevent network stealing: namely, it produces the same responses and classification accuracy, but it is extremely sensitive to weight changes.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
